[ICoN: Integration using Co-attention across Biological Networks

Abstract

AbstractMotivationMolecular interaction networks are powerful tools for studying cellular functions. Integrating diverse types of networks enhances performance in downstream tasks such as gene module detection and protein function prediction. The challenge lies in extracting meaningful protein feature representations due to varying levels of sparsity and noise across these heterogeneous networks.ResultsWe propose ICoN, a novel ‘co-attention’-based, denoising, unsupervised graph neural network model that takes multiple protein-protein association networks as inputs and generates an integrated single network by computing a unified feature representation for each protein. A key contribution of ICoN is a novel approach that enables cross-network communication through co-attention during training. The model also incorporates a denoising training technique, introducing perturbations to each input network and training the model to reconstruct the original network from its corrupted version, a method previously unexplored in network integration.Our experimental results demonstrate that ICoN surpasses individual networks across three downstream tasks: gene module detection, gene coannotation prediction, and protein function prediction. Compared to existing unsupervised network integration models, ICoN exhibits superior performance across the majority of downstream tasks and exhibits enhanced robustness against noise. This work introduces a promising approach for effectively integrating diverse protein-protein association networks, aiming to achieve a biologically meaningful unified representation of proteins.AvailabilityThe ICoN software is available under the GNU Public License v3 athttps://github.com/Murali-group/ICoN.